Hypothesis-only classification reaches 64% accuracy on SNLI, revealing dataset biases in SNLI and MultiNLI that the authors quantify and propose a simple mitigation for.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Investigating Biases in Textual Entailment Datasets
Hypothesis-only classification reaches 64% accuracy on SNLI, revealing dataset biases in SNLI and MultiNLI that the authors quantify and propose a simple mitigation for.